© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 31116
    Prev
    Next
    INDEX
    Explanations

    movement into or around

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    /on
    -0.11
    /from
    -0.10
    æĸ¹æīį
    -0.10
    并对
    -0.09
    ylon
    -0.08
    èĥ´
    -0.08
    éĿ¢è²Į
    -0.08
    就好
    -0.08
    æĪij们è¦ģ
    -0.08
    åŀĤ
    -0.08
    POSITIVE LOGITS
    åĩºæĿ¥
    0.12
     into
    0.11
    æĬ±
    0.11
    è¿ĽæĿ¥
    0.11
    åħ¥å¢ĥ
    0.11
    è¿Ľåİ»
    0.10
     INTO
    0.10
     across
    0.10
    è¿Ľåħ¥
    0.10
     Ritch
    0.09
    Activations Density 0.177%

    No Known Activations